Andrew is a Senior Software Engineer at Cloudsoft and the founder of the Clocker project. He is a contributor to several Open Source projects including jclouds and Qpid and is on the Apache Brooklyn PMC. Areas of interest include Distributed Systems, Virtualisation, Messaging, Information Security and LOLcats. Prior to joining Cloudsoft, Andrew worked for various investment banks as a Software Engineer and Security Consultant and has over twenty years experience in the IT industry.
Anomaly Detection: A contradiction with lots of potential – Pieter Buteneers
Anomaly detection. Many people speak about it, some use it, some build it, some love it but many hate it. What is this strange beast and why should you use it? In this talk I will go through my experiences in developing an anomaly detection system for CoScale. I will discuss many of the methods I have used, tested and dismissed. What works? What doesn’t work? And why doesn’t it work? I will go into detail about the difference between statistical significance and relevance. Why the univariate approach is the way to go. Why most multivariate techniques don’t work. And why you should keep it simple, stupid. Since the world of anomaly detection is filled with contradictions, I will show why I am wrong to claim that multivariate techniques don’t work. When do they outperform univariate techniques? And why should you keep it simple, stupid once again? Contrary to my advice, I know some of you will try the more complex or ‘less stupid’ techniques. And you should! But not before you know how you can use them without them misleading you.
Storage for containers shouldn’t be annoying – Karolis Rusenas
All is good about microservices, deploying your applications and databases through containers but when it comes to persistent storage – you usually end up using something that your cloud provider gives you (mounting an EBS volume) which helps but doesn’t really solve the problem since it’s usually clunky, expensive and locks you into some particular cloud provider. Sometimes you can’t even access those public cloud features (i.e. you are under your corporate’s iron dome) . Thankfully, StorageOS (do not pronounce it as Storageous) comes to the rescue! The idea is to run a masterless & distributed system on containers next to your applications that takes care of persisting and backing your data while preserving high speed (or the slowness if you are using some ancient hardware) of local disks. This system is built with Go, C, Consul and Nats, part of my talk will touch architecture and why we chose some things over the others. StorageOS is also very easy to use – docker plugin,GUI, HTTP API
or CLI. There are rumours that even project managers were able to provision volumes via graphical user interface. People, who have not attended our demo during the event, will be sold to circus.
#nostaging – Pavel Chunyayev
In the fasted-paced modern world we can’t afford having a staging anymore. It slows things down and gives a fake feeling of safeness.
Introducing Cloudsoft Container Service – Andrew Kennedy
Learn how to deploy and manage a production-quality Docker Swarm deployment with all the bells and whistles using the latest Cloudsoft AMP 4.0.0, then see how the included Cloudsoft Container Service is able to orchestrate your business applications, using both containers and virtual machines to take full advantage of the resources and constraints available. Containers are awesome, but they need to play well with others, particularly the cloud. With CCS you can design and build hybrid applications spanning both worlds, and roll them out into managed or unmanaged, public or private infrastructure. We will give a very short demonstration showing how CCS can deploy a global Hyperledger Blockchain, and then execute transactions there.